1. Field of the Invention
The present invention relates to an image processing device and an image processing method for processing a three-dimensional image formed by a plurality of tomographic images, which are obtained by tomographic imaging of a subject, as well as a program for causing a computer to carry out the image processing method.
2. Description of the Related Art
In recent years, in the field of medical images, modalities using various technologies, such as an X-ray CT (Computed Tomography) system, an ultra-sound (US) diagnosis system, an MRI (Magnetic Resonance Imaging) system, a PET (Positron Emission Tomography) system, and an SPET (Single-Photon Emission Tomography) system are used, besides an X-ray imaging system. Along with improvement of speed and performance, such as to accommodate multislice technology, of such modalities, it has become possible to image two or more parts of a subject in a series of imaging to obtain several hundreds to several thousands of high-resolution tomographic images. However, it takes time to observe these tomographic images one by one. Further, it requires a trained doctor to understand a three-dimensional shape of a structure to be observed (an organ or a tissue, such as body surface, bone, heart, lung field or liver) only from the tomographic images.
To address this problem, various techniques have been proposed to enhance visibility of the entire structure, and further a lesion contained in the structure, by recognizing a structure of interest, and generating, from tomographic images containing the structure of interest, a three-dimensional image of the structure of interest by analyzing the tomographic images with using a method, such as maximum intensity projection (MIP) or minimum intensity projection (MinIP), to achieve MIP display, or to achieve volume rendering (VR) display of the three-dimensional image.
On the other hand, there are a wide variety of analysis applications for performing such an analysis. Further, besides the analysis applications, there are a wide variety of color templates for changing a shading pattern of a VR display of a three-dimensional image, image processing applications and display applications (which will hereinafter be referred to as analysis applications, etc.). Therefore, the number of menus for selecting the analysis applications, etc., during image interpretation is increased, and a burdensome operation is imposed on the user, such as a reading physician or a technologist, for selecting menu items.
A technique which involves: calculating index values which indicate anatomical characteristics from a medical image; determining whether the imaged subject is an adult, infant, young child or child based on the calculated index values; selecting optimal imaging menu items to be selected from imaging menu items depending on the subject; and displaying only the selected imaging menus has been proposed (see Japanese Patent No. 2007-185429, which will hereinafter be referred to as Patent Document 1). According to the technique disclosed in Patent Document 1, when a medical image is displayed, unnecessary imaging menu items for the imaged subject are not displayed, and thus a burden imposed on the user for selecting the imaging menu items can be reduced.
In the case where tomographic imaging is carried out using the above-described modality, such as a CT system, there are often the cases where two or more body parts are imaged, such as from the chest to the abdomen or from the head to the chest, besides the cases where only a single body part (for example, only the chest or abdomen) is imaged, in a single examination. The above-described analysis applications, etc., are prepared for each body part. Therefore, when two or more body parts are imaged, two or more of the analysis applications, etc., are used for analysis, or the like. Although the technique disclosed in Patent Document 1 allows selection of an optimal imaging menu item depending on the subject, the technique is not applicable to the cases where tomographic images containing two or more parts are obtained, as described above.